Text splitter
Split text by characters, words, lines, a delimiter or a regular expression, with break-at-punctuation, keep-delimiter, trimming and numbering options plus chunk statistics.
Runs in your browserEvery computation happens in your browser — your data never leaves this device.
Statistics
724147403Chunks
UniKit is a purely client-side toolbox —
every calculation runs inside your brow
ser. Split text by characters, words, l
ines, a delimiter or a regular expressio
n. The chunks are listed one by one so
you can copy a single chunk or all of th
em.
What this tool does
- Chop long copy into bite-sized pieces before feeding it to a translator, an LLM or a subtitle tool — split by characters, words, lines, a delimiter or a regex.
- With character splitting, enable "break at punctuation or spaces" so the cut lands near a comma or full stop instead of the middle of a phrase.
- Regex splitting is ideal for paragraphs: \n\s*\n cuts an entire document on blank lines in one go.
- Copy chunks one by one or all at once (blank line between chunks), and number them when you need to reference a specific piece.
Example
Input
UniKit 是一个即开即用的在线工具箱，所有计算都在浏览器里完成。
Output
1. UniKit 是一个即开即用的在线工具箱， 2. 所有计算都在浏览器里完成。
This is "20 characters per chunk" with break-at-punctuation, trimming and numbering enabled: the first chunk ends after the comma because that is the nearest break point.
Frequently asked questions
Can character splitting cut an emoji in half?
No. The character mode counts Unicode code points (iterating with Array.from), so a surrogate-pair emoji counts as one character and is never split into two broken halves.
How does "break at punctuation or spaces" pick the cut point?
It computes the target position, then walks backwards from there — at most to the halfway point — and breaks after a comma, full stop, question mark, space or closing quote. A chunk may therefore be slightly shorter than the number you typed, but the break reads naturally.
What is the difference between keeping and dropping the delimiter?
Dropped, the delimiter disappears from the output; kept, it is prepended to the next chunk. Enable it if you want to concatenate the chunks back into the original. In regex mode the tool wraps a pattern without capture groups in parentheses to make this work.
Why do I get 0 chunks, or only one?
An empty input has nothing to split, and a single chunk means the whole text still fits within the per-chunk length you set. Check the length value or switch to another split mode.
Keywords：split texttext splitterchunkdelimiterregex文本拆分分段按字符拆分按分隔符正则拆分